university 2/n
Chapter 1.Introduction
of register,and register is the‘expression-plane’of genre;register is in turn the
‘content-plane’of language.Lee(2001)surveys the use of these terms,providing
one of the most comprehensive discussions of how they have been used in previous
research(as well as terms like text type and style).
When research studies have attempted to distinguish between register and
genre(such as Ventola 1984;Martin 1985;Couture 1986;Swales 1990;Ferguson
1994),the distinction has been applied at two different levels of analysis:
1.to the object of study;
2.to the characteristics of language and culture that are investigated.
With regard to the first–the object of study–the term register(when it is distin-
guished from genre)has been used to refer to a general kind of language associated
with a domain of use,such as a‘legal register’,‘scientific register’,or‘bureaucratic
register’.In contrast,the term genre has been used to refer to a culturally recog-
nized‘message type’with a conventional internal structure,such as an affidavit,a
biology research article,or a business memo.However,it is difficult to pin down
specific differences in the object of study referred to by the two terms;both are
used in reference to linguistic varieties associated with particular situations of use
and particular communicative purposes.For example,Ferguson(1994)notes that
‘cookbook recipes’can be studied both as a register and as a genre.
With regard to the second–the characteristics of language and culture that are
described–register studies have focused on lexico-grammatical features,show-
ing how the use of particular words,word types,and grammatical features vary
systematically in accord with the situation of use(factors such as interactivity,
personal involvement,mode,production circumstances,and communicative pur-
pose).In contrast,genre studies have usually focused on socio-cultural actions;for
example,genres are“how things get done,when language is used to accomplish
them”(Martin 1985:250),and“frames for social action”(Bazerman 1997:19).
As a result,genre studies have often been concerned with issues of ideology and
social power.
The analytical approach in the present book incorporates elements from pre-
vious conceptions of both‘register’and‘genre’:The objects of study in the present
book are culturally-recognized varieties with conventional internal structure,like
office hours,service encounters,textbooks,and course syllabi.However,the level
of analysis focuses on characteristic lexico-grammatical linguistic features,rather
than discussions of discourse communities and issues of ideology and power.
I use the term register in the present study,as in my previous recent studies
(e.g.,Biber 1995;Biber,Conrad,&Reppen 1998;Biber et al.1999;Conrad&Biber
2001),to refer to situationally-defined varieties described for their characteristic
lexico-grammatical features.Thus,register is a general cover term,with no implied
theoretical distinction to‘genre’as it has been used in some other studies:??University Language
Register is used as a cover term for any language variety defined in terms of a par-
ticular constellation of situational characteristics.That is,register distinctions are
defined in non-linguistic terms,including the speaker’s purpose in communica-
tion,the topic,the relationship between speaker and hearer,and the production
circumstances.However,as illustrated by the papers in this book,there are usually
important linguistic differences across registers that correspond to the differences
in situational characteristics.
In many cases,registers are named varieties within a culture,such as nov-
els,biographies,letters,memos,book reviews,editorials,sermons,lectures,and
debates.However,registers can be defined at any level of generality,and more spe-
cialized registers may not have widely used names.For example,‘academic prose’
is a very general register,while‘methodology sections in experimental psychology
articles’is a much more highly specified register.
Because registers vary in the extent to which they are specified situationally,the
texts within registers also vary in the extent to which their linguistic characteristics
are similar.At one extreme,texts from a specialized register(such as methodology
sections of experimental psychology articles or air-traffic-controller talk)tend to
be very similar in their linguistic characteristics,corresponding to the extent to
which the register is specified situationally.In contrast,the texts in a general reg-
ister,such as academic prose or fiction,tend to exhibit a wide range of linguistic
variation.(Biber&Conrad 2001:3)
?.?.?Previous Multi-Dimensional studies of academic registers
Multi-Dimensional studies have attempted to provide comprehensive linguis-
tic descriptions of academic registers in comparison to other registers.Multi-
Dimensional(MD)analysis identifies the salient linguistic co-occurrence patterns
in a language–the‘dimensions’–and then spoken and written registers are
compared in the linguistic space defined by those dimensions.The analysis is com-
prehensive in that a wide range of linguistic features is analyzed in each text(see
Biber et al.2003).Those features are reduced to a few underlying dimensions,
where each dimension represents a set of individual features that tend to co-occur
in texts.Registers are then described and compared with respect to the underlying
dimensions,rather than with respect to individual features.A fuller description of
the methodology is given in Chapter 8 below.
MD analysis was originally developed to study the overall patterns of variation
among spoken and written registers in English(Biber 1985,1986,1988).Academic
prose was one of the registers included in those analyses,described in compari-
son to 24 other spoken and written registers.Biber(1988:180–195)additionally
provides more detailed MD descriptions of the variation among selected subdis-
ciplines within academic prose(e.g.,Natural Science,Medicine,Social Science,Chapter 1.Introduction
etc.).Grabe(1987)also uses MD analysis to study the overall patterns of variation
among academic disciplines,in comparison to other written registers.
Building on those early studies,a number of subsequent MD investigations
have focused on the characteristics of particular academic registers.For example,
Atkinson(1992)and Biber and Finegan(1994c)focus on the MD characteristics
of medical research articles in relation to other registers.Atkinson(1996,1999)
provides a detailed description of scientific research articles,using MD analysis
to trace the historical evolution of this register,and interpreting the linguistic
patterns relative to changes in rhetorical structure and function.
Conrad(1996,2001)provides a detailed MD comparison of research arti-
cles and textbooks from two major academic disciplines:ecology and American
history.Carkin(2001)also analyzes the MD characteristics of textbooks in com-
parison to classroom teaching,focusing on introductory undergraduate courses
in biology and macroeconomics.Csomay(2000)surveys the MD characteristics
of classroom lectures.Finally,Biber,Conrad,Reppen,Byrd,and Helt(2002,2003)
apply MD analysis to compare a wide range of spoken and written university regis-
ters,using the same corpus as in the present study(see Chapter 2 below).Chapter
8 in the present book extends this line of research,considering the MD charac-
teristics of academic prose in relation to the range of other spoken and written
university registers.
?.?.?The grammatical description of written academic prose in The Longman
Grammar of Spoken and Written English
Most previous studies of academic language have described the functions of a
particular linguistic feature in a particular register.The research goal has been to
document the distinctive uses of the target linguistic feature in that register,rather
than to provide a comprehensive linguistic description of the register.
One major exception to this trend is the Longman Grammar of Spoken and
Written English(LGSWE;Biber,Johansson,Leech,Conrad,&Finegan 1999),
which provides a comprehensive linguistic description of academic prose.The
LGSWE is a corpus-based reference grammar of English;it describes the range
of grammatical features in English and compares the use of these features in four
major registers:conversation,fiction,newspapers,and academic prose.The reg-
ister comparisons are based on analysis of a representative corpus of texts(the
Longman Spoken and Written English Corpus)containing approximately 5 mil-
lion words from each register(see LGSWE,pp.24–35).The LGSWE describes the
structural characteristics of grammatical features in English,but at the same time,
it describes the patterns of use for those features:register differences and other
contextual factors that influence the patterns of variation.For this reason,the??University Language
LGSWE provides relatively comprehensive descriptions of the four registers.The
description of academic prose is especially relevant for our purposes here.
The academic prose subcorpus used for the LGSWE consists of both academic
books(2.65 million words)and academic research articles(2.68 million words;see
LGSWE,pp.32–34).Texts from both books and research articles were collected
from a number of academic disciplines,including biology,chemistry,medicine,
sociology,education,and law/history/politics.The corpus thus represents aca-
demic prose as a general register.
The LGSWE presents many frequency findings that describe the patterns of
use for particular grammatical features.Table 1.1 lists the major‘academic fea-
tures’identified in the LGSWE:features that occur much more frequently in
academic prose than in the other registers.For example,corpus analysis in the
LGSWE shows that there are about 300,000 nouns per million words in academic
prose,compared to only around 150,000 per million words in conversation(p.
235).Similarly,there are about 80,000 adjectives per million words in academic
prose versus only 20,000 per million words in conversation(and around 60,000
per million words in fiction and newspapers;p.506).Thus,these grammatical
features can be considered characteristic of academic prose by virtue of their dis-
tribution:although they can be used in all registers,they turn out to be much more
common in academic prose.
Taken together,the findings reported in the LGSWE show that‘academic’
grammatical features come from most structural categories.Three word classes are
especially prevalent:nouns,adjectives,and prepositions.Overall,these grammat-
ical classes are more frequent in academic prose than in other registers,and there
are many related specific features that are especially characteristic of academic
prose(e.g.,nominalizations,noun phrases with multiple modifiers,stance noun
+of-phrase).In contrast,verbs overall are much less common in academic prose
than in other registers,although there are specific verb categories that are typi-
cal of academic prose(e.g.,copula be,existence verbs,derived verbs,and passive
voice verbs).Similarly,there are specific categories of adverbs and adverbials(e.g.,
linking adverbials)that are especially common in academic prose,even though
adverbs overall are more common in spoken registers.
It is difficult to make any global generalizations for dependent clauses.Rather,
specific clause types have their own distributions.For example,finite relative
clauses are much more common in writing than in conversation,but they are most
common in newspaper writing and fiction rather than academic prose.However,
relative clauses with the relative pronoun which are most frequent in academic
prose.Non-finite relative clauses(-ing clauses and-ed clauses)are also especially
common in academic prose(and newspaper writing).Many complement clause
types are actually much more common in conversation than academic prose,es-
pecially that-clauses controlled by verbs(e.g.,I know that he did it).Only oneChapter 1.Introduction
Table 1.1 Grammatical features that are especially common in academic prose(based on a
survey of the Longman Grammar of Spoken and Written English)
Feature Pattern of use
Nouns and noun phrases:
Nouns:overall
p.65
Approximately 60%of all content words in academic prose are
nouns
Nouns vs.pronouns
pp.235–236
Nouns are much more common than pronouns in academic
prose,especially in object positions
Absence of pronouns
pp.235–236
Pronouns are generally rare in academic prose
Specific pronouns:this and
generic one
pp.349–350,354–355
Much more common in academic prose;this is used for
immediate textual reference;one is used for generic rather than
specific reference
Plural nouns
pp.291–292
Much more common in writing than in conversation;most
common in academic prose
Nominalizations
pp.322–323
Much more common in academic prose,especially nouns formed
with-tion and-ity(e.g.distribution,similarity)
Anaphoric expressions
pp.237–238
Anaphoric reference is usually expressed with a determiner+
noun(rather than a pronoun)
Definite article the
pp.267–269
Much more common in writing than in conversation;most
common in academic prose
Demonstrative determiners
pp.270,274–275
Most common in academic prose;especially this and these
Noun phrases with
modifiers
p.578
60%of all noun phrases in academic prose have a modifier
Noun phrases with
pre-modifiers
pp.589,597
Very common in academic prose(and newspapers)
Nouns as pre-modifiers
pp.589–596
Very common in academic prose(and newspapers)(e.g.,
government agencies)
Noun phrases with
post-modifiers
pp.606–608
Very common in academic prose(and newspapers)
Noun phrases with
multiple post-modifiers
pp.640–644
Most common in academic prose(e.g.,the utilization of such
devices for social purposes)
Noun and/or noun
binomial phrases
pp.1033–1034
Most common in academic prose(e.g.,size and shape)
Adjectives and adjective
phrases:
Adjectives:overall
pp.65,506
Adjectives are much more common in academic prose than in
conversation or fiction
Attributive adjectives
pp.506,589
Much more common in academic prose(e.g.,the basic logical
content)??University Language
Table 1.1(continued)
Feature Pattern of use
Specific predicative
adjectives
p.440
Several predicative adjectives are notably more common in
academic prose than in other registers:different,important,
difficult,possible,necessary,available,useful
Derived adjectives
pp.531–533
Much more common in academic prose,especially adjectives
formed with-al(e.g.functional,regional)
Verbs and verb phrases:
Copula be and copular verb
become
pp.359–360,437–439
Most common in academic prose
“Existence”verbs
pp.366,369,419
Much more common in writing than in conversation;Most
common in academic prose(e.g.,include,involve,indicate)
Specific lexical verbs
pp.367–372
Several verbs are notably more common in academic prose than
in other registers:
Activity verbs:use,produce,provide,apply,form,obtain,reduce
Communication verbs:describe,suggest
Mental verbs:consider,assume,determine
Causative/Occurrence/Existence verbs:follow,allow,require,
include,involve,contain,exist,indicate,represent
Specific prepositional verbs
pp.416–418
Several prepositional verbs are notably more common in
academic prose than in other registers:
Activity verbs:deal with,BE applied to,BE used in,BE derived from
Communication verbs:refer to
Mental verbs:BE known as
Causative/Occurrence/Existence verbs:lead to,result in,occur
in,depend on,consist of,BE based on,BE associated with,BE
related to
Verbs with inanimate
subjects
pp.378–380
Common only in academic prose(e.g.,such comparisons
suggest...)
Derived verbs
pp.400–403
Most common in academic prose,especially verbs formed with
re-and-ize(e.g.redefine,computerize)
Tense and aspect
pp.456–462
Academic prose relies primarily on simple aspect,present tense
verb phrases
Passive voice
pp.476–480,937–940
Much more common in academic prose,especially the‘short’
passive(with no by-phrase)
Specific passive verbs
pp.478–480
Several verbs are especially common with passive voice in
academic prose;for example:BE+made,given,taken,used,
found,seen,considered,shown
Adverbs and adverbials:
Specific adverbs
pp.560–563
Several adverbs are notably more common in academic prose
than in conversation:often,usually,significantly,more,relatively,
especially,particularly,generally,indeed
Specific amplifiers
pp.560–563
A few amplifiers are notably more common in academic prose
than in conversation:extremely,highlyChapter 1.Introduction
Table 1.1(continued)
Feature Pattern of use
Specific degree adverbs
pp.566–569
A few degree adverbs are notably more common in academic
prose than in conversation:relatively,fairly,slightly
Linking adverbials
pp.766,880–882
Most common in academic prose;especially
however,thus,therefore,for example(e.g.)
Purpose and concessive
adverbials
pp.784,786,820–826
Most common in academic prose(e.g.,in order to,although)
Dependent clause features:
Relative clauses with the
relative pronoun which
pp.609–612
Most common in academic prose
Participle clauses as
post-modifiers in noun
phrases
pp.606,630–632
Very common in academic prose(and newspapers)(e.g.,the
assumptions given above)
Noun complement clauses
as post-modifiers in noun
phrases
pp.645–655
Very common in academic prose(and newspapers)(e.g.,the fact
that...;the attempt to...)
Abstract noun+of+
ing-clause
pp.653–655
Most common in academic prose,especially with the head nouns
way,cost,means,method,possibility,effect,problem,process,risk
(e.g.,methods of assessing error)
Extraposed that-clauses
pp.672–675
Most common in academic prose,especially controlled by the
adjectives clear,(un)likely,and(im)possible
Extraposed to-clauses
pp.672–675
Most common in academic prose,especially controlled by
adjectives(e.g.,(im)possible,difficult,hard,important,necessary)
Subject predicative
to-clause
pp.714–715,723
Common only in academic prose(and newspapers)(e.g.,The first
step is to evaluate the expression)
ing-clauses controlled by
adjective predicates
p.749
Most common in academic prose;(e.g.,capable of,important
for/in,useful for/in:formalist strategies are useful for analyzing
drama)
Concessive adverbial
clauses
pp.820–825
Most common in academic prose(and newspapers)(though,
although)
Other features
Prepositions
p.92
Most common in academic prose
Of-phrases
pp.301–302
Much more common in writing than in conversation;
Most common in academic prose
Prepositional phrases as
post-modifiers in noun
phrases
p.606–608,634–638
Very common in academic prose(and newspapers)(e.g.,the effect
on the final state)??University Language
Table 1.1(continued)
Feature Pattern of use
Stance noun+of-phrase
pp.984–986
Most common in academic prose,especially possibility of,value of,
importance of,problem of,understanding of
that/those+of-phrase
pp.307–308
Common only in academic prose
Preposition+which in
relative clauses with
adverbial gaps
pp.624–626
Common only in academic prose,especially in which and to which
Selected coordination tags:
and so on,etc.
pp.116–117
Common only in academic prose
Quantifier each Most common in academic prose
Semi-determiners same,
other,certain,and such
pp.282–283
Much more common in academic prose
Dual gender reference:he
or she,his or her,he/she
pp.316–317
Common only in academic prose
Lexical bundles with noun
phrases and/or
prepositional phrases
pp.997,1015–1019
Very common in academic prose;e.g.,the end of the,the nature of
the,one of the most,the way in which,the extent to which,the fact
that the,as a result of,at the time of,in the
case/absence/form/presence of on the basis of,on the other hand
major type of complement clause is especially characteristic of academic prose:
extraposed clauses controlled by adjectives(e.g.,It is possible that...,It is impor-
tant to...).(Subject predicative to-clauses are also found primarily in academic
prose,although they are not especially frequent overall.)Finally,adverbial clauses
are most common overall in fiction.Even conversation uses adverbial clauses to a
slightly greater extent than academic prose.Only one sub-type of adverbial clause
is especially frequent in academic prose:concessive clauses.
In summary,there are few general linguistic features that are uniquely char-
acteristic of academic prose.The most distinctive features of academic prose are
specific grammatical features associated with a particular set of words,such as ex-
traposed complement clauses controlled by stance adjectives.However,a much
larger set of features–such as nouns and prepositional phrases–occur to some
extent in every register;these features can be considered‘academic’because they
are especially common in academic prose.Chapter 1.Introduction
?.?Strengths and weaknesses of previous research
As the survey in the preceding sections show,there have been numerous studies
on academic language over the past two decades.Most of these studies focus on
the use of a particular word or linguistic feature in a particular academic register,
such as the hedge kind of in geology lectures.Taken together,these studies provide
a useful foundation for the study of linguistic similarities and differences among
university registers.
From a structural perspective,previous studies have described features at
many different linguistic levels:words(e.g.,thing),extended collocations and
formulaic language(e.g.,come into play),part-of-speech classes(e.g.,modals),
grammatical/functional classes(e.g.,hedges),syntactic constructions(e.g.,noun
phrase modifiers,conditional adverbial clauses),and overall discourse organiza-
tion.In most studies,linguistic features are described relative to their discourse
functions,such as:conveying informational content;expressing hedging,evalu-
ation,and stance;and signaling the discourse organization(topics,background
context,informational focus or packaging,etc.).In addition,most studies focus on
a particular register and academic discipline,such as biology textbooks or medical
research articles.
Despite the large number of previous studies,there are still some obvious gaps
in what we currently know about university language.First,we lack comprehensive
linguistic descriptions of most university registers.While written academic prose
has been thoroughly investigated,we do not have comprehensive linguistic de-
scriptions for other academic/university registers.Textbooks and academic speech
are probably the best studied of these other registers,with several studies describ-
ing the functions of particular linguistic features.Taken together,these studies
identify some of the salient features of those registers,but they do not amount
to a comprehensive linguistic description of either textbooks or academic speech.
And we know even less about registers like office hours,study groups,and institu-
tional writing(e.g.,university catalogs),because there have been few studies of any
kind based on texts from those registers.In addition,apart from previous Multi-
Dimensional studies,we know little about the overall patterns of variation among
university registers.The T2K-SWAL Project,introduced in the following section,
was designed to help fill these gaps.
?.?Background of the present book:Introduction to the T2K-SWAL Project
The present book grew out of the TOEFL 2000 Spoken and Written Academic
Language(T2K-SWAL)Project,sponsored by the Educational Testing Service(see
Biber et al.2004).The project included three major components:??University Language
–the construction of a large corpus of spoken and written university registers
(the T2K-SWAL Corpus);
–the description of language use in the university based on extensive linguistic
analysis of that corpus;
–the development of analytical tools to describe the linguistic characteristics of
exam prompts relative to the corpus.
The initial motivation for the project grew out of the need for an external standard
to evaluate the representativeness of ESL/EFL materials.That is,given the lack of
a comprehensive description of university language,it has been nearly impossible
to evaluate the extent to which textual materials for ESL/EFL instruction and as-
sessment actually represent the language of the target registers.Specifically in the
context of the TOEFL 2000 effort undertaken by ETS(see Jamieson et al.2000),
we lacked the tools to determine whether the texts used in listening and reading
portions of the TOEFL exam accurately represent the linguistic characteristics of
spoken and written academic registers(see Enright et al.2000;Bejar et al.2000).
One reason for this gap is that there have been no readily available text corpora
of university language that could be used for research studies.The T2K-SWAL
Corpus was compiled to meet this need.Then,based on analysis of that corpus,
the T2K-SWAL project focused on two primary goals:
–to identify the salient patterns of language use found in each university register
(and across disciplines,levels,etc.);
–to develop diagnostic tools and procedures for assuring that the language used
in TOEFL Listening Comprehension and Reading Comprehension tasks were
representative of actual language use in the university.
Biber et al.(2004)is a technical report describing the T2K-SWAL Project and the
design,construction,and linguistic analysis of the T2K-SWAL Corpus.
The T2K-SWAL Project was a large-scale study,involving the collaborative ef-
forts of multiple researchers.Although the project was coordinated at Northern
Arizona University,and most linguistic analyses were carried out there,the col-
lection of texts for the corpus was supervised by four co-principal investigators
at different universities:Susan Conrad at Iowa State University;Randi Reppen
at NAU;Pat Byrd at Georgia State University;and Marie Helt at California State
University,Sacramento.In addition,there were numerous research assistants and
student workers from all four universities who helped collect,transcribe,scan,tag,
tag-edit,and analyze the corpus.
Work on the T2K-SWAL Project proceeded in three major stages:(1)design,
construction,and grammatical“tagging”of the T2K-SWAL Corpus;(2)linguistic
analysis of the patterns of register variation in the corpus;and(3)development of
of register,and register is the‘expression-plane’of genre;register is in turn the
‘content-plane’of language.Lee(2001)surveys the use of these terms,providing
one of the most comprehensive discussions of how they have been used in previous
research(as well as terms like text type and style).
When research studies have attempted to distinguish between register and
genre(such as Ventola 1984;Martin 1985;Couture 1986;Swales 1990;Ferguson
1994),the distinction has been applied at two different levels of analysis:
1.to the object of study;
2.to the characteristics of language and culture that are investigated.
With regard to the first–the object of study–the term register(when it is distin-
guished from genre)has been used to refer to a general kind of language associated
with a domain of use,such as a‘legal register’,‘scientific register’,or‘bureaucratic
register’.In contrast,the term genre has been used to refer to a culturally recog-
nized‘message type’with a conventional internal structure,such as an affidavit,a
biology research article,or a business memo.However,it is difficult to pin down
specific differences in the object of study referred to by the two terms;both are
used in reference to linguistic varieties associated with particular situations of use
and particular communicative purposes.For example,Ferguson(1994)notes that
‘cookbook recipes’can be studied both as a register and as a genre.
With regard to the second–the characteristics of language and culture that are
described–register studies have focused on lexico-grammatical features,show-
ing how the use of particular words,word types,and grammatical features vary
systematically in accord with the situation of use(factors such as interactivity,
personal involvement,mode,production circumstances,and communicative pur-
pose).In contrast,genre studies have usually focused on socio-cultural actions;for
example,genres are“how things get done,when language is used to accomplish
them”(Martin 1985:250),and“frames for social action”(Bazerman 1997:19).
As a result,genre studies have often been concerned with issues of ideology and
social power.
The analytical approach in the present book incorporates elements from pre-
vious conceptions of both‘register’and‘genre’:The objects of study in the present
book are culturally-recognized varieties with conventional internal structure,like
office hours,service encounters,textbooks,and course syllabi.However,the level
of analysis focuses on characteristic lexico-grammatical linguistic features,rather
than discussions of discourse communities and issues of ideology and power.
I use the term register in the present study,as in my previous recent studies
(e.g.,Biber 1995;Biber,Conrad,&Reppen 1998;Biber et al.1999;Conrad&Biber
2001),to refer to situationally-defined varieties described for their characteristic
lexico-grammatical features.Thus,register is a general cover term,with no implied
theoretical distinction to‘genre’as it has been used in some other studies:??University Language
Register is used as a cover term for any language variety defined in terms of a par-
ticular constellation of situational characteristics.That is,register distinctions are
defined in non-linguistic terms,including the speaker’s purpose in communica-
tion,the topic,the relationship between speaker and hearer,and the production
circumstances.However,as illustrated by the papers in this book,there are usually
important linguistic differences across registers that correspond to the differences
in situational characteristics.
In many cases,registers are named varieties within a culture,such as nov-
els,biographies,letters,memos,book reviews,editorials,sermons,lectures,and
debates.However,registers can be defined at any level of generality,and more spe-
cialized registers may not have widely used names.For example,‘academic prose’
is a very general register,while‘methodology sections in experimental psychology
articles’is a much more highly specified register.
Because registers vary in the extent to which they are specified situationally,the
texts within registers also vary in the extent to which their linguistic characteristics
are similar.At one extreme,texts from a specialized register(such as methodology
sections of experimental psychology articles or air-traffic-controller talk)tend to
be very similar in their linguistic characteristics,corresponding to the extent to
which the register is specified situationally.In contrast,the texts in a general reg-
ister,such as academic prose or fiction,tend to exhibit a wide range of linguistic
variation.(Biber&Conrad 2001:3)
?.?.?Previous Multi-Dimensional studies of academic registers
Multi-Dimensional studies have attempted to provide comprehensive linguis-
tic descriptions of academic registers in comparison to other registers.Multi-
Dimensional(MD)analysis identifies the salient linguistic co-occurrence patterns
in a language–the‘dimensions’–and then spoken and written registers are
compared in the linguistic space defined by those dimensions.The analysis is com-
prehensive in that a wide range of linguistic features is analyzed in each text(see
Biber et al.2003).Those features are reduced to a few underlying dimensions,
where each dimension represents a set of individual features that tend to co-occur
in texts.Registers are then described and compared with respect to the underlying
dimensions,rather than with respect to individual features.A fuller description of
the methodology is given in Chapter 8 below.
MD analysis was originally developed to study the overall patterns of variation
among spoken and written registers in English(Biber 1985,1986,1988).Academic
prose was one of the registers included in those analyses,described in compari-
son to 24 other spoken and written registers.Biber(1988:180–195)additionally
provides more detailed MD descriptions of the variation among selected subdis-
ciplines within academic prose(e.g.,Natural Science,Medicine,Social Science,Chapter 1.Introduction
etc.).Grabe(1987)also uses MD analysis to study the overall patterns of variation
among academic disciplines,in comparison to other written registers.
Building on those early studies,a number of subsequent MD investigations
have focused on the characteristics of particular academic registers.For example,
Atkinson(1992)and Biber and Finegan(1994c)focus on the MD characteristics
of medical research articles in relation to other registers.Atkinson(1996,1999)
provides a detailed description of scientific research articles,using MD analysis
to trace the historical evolution of this register,and interpreting the linguistic
patterns relative to changes in rhetorical structure and function.
Conrad(1996,2001)provides a detailed MD comparison of research arti-
cles and textbooks from two major academic disciplines:ecology and American
history.Carkin(2001)also analyzes the MD characteristics of textbooks in com-
parison to classroom teaching,focusing on introductory undergraduate courses
in biology and macroeconomics.Csomay(2000)surveys the MD characteristics
of classroom lectures.Finally,Biber,Conrad,Reppen,Byrd,and Helt(2002,2003)
apply MD analysis to compare a wide range of spoken and written university regis-
ters,using the same corpus as in the present study(see Chapter 2 below).Chapter
8 in the present book extends this line of research,considering the MD charac-
teristics of academic prose in relation to the range of other spoken and written
university registers.
?.?.?The grammatical description of written academic prose in The Longman
Grammar of Spoken and Written English
Most previous studies of academic language have described the functions of a
particular linguistic feature in a particular register.The research goal has been to
document the distinctive uses of the target linguistic feature in that register,rather
than to provide a comprehensive linguistic description of the register.
One major exception to this trend is the Longman Grammar of Spoken and
Written English(LGSWE;Biber,Johansson,Leech,Conrad,&Finegan 1999),
which provides a comprehensive linguistic description of academic prose.The
LGSWE is a corpus-based reference grammar of English;it describes the range
of grammatical features in English and compares the use of these features in four
major registers:conversation,fiction,newspapers,and academic prose.The reg-
ister comparisons are based on analysis of a representative corpus of texts(the
Longman Spoken and Written English Corpus)containing approximately 5 mil-
lion words from each register(see LGSWE,pp.24–35).The LGSWE describes the
structural characteristics of grammatical features in English,but at the same time,
it describes the patterns of use for those features:register differences and other
contextual factors that influence the patterns of variation.For this reason,the??University Language
LGSWE provides relatively comprehensive descriptions of the four registers.The
description of academic prose is especially relevant for our purposes here.
The academic prose subcorpus used for the LGSWE consists of both academic
books(2.65 million words)and academic research articles(2.68 million words;see
LGSWE,pp.32–34).Texts from both books and research articles were collected
from a number of academic disciplines,including biology,chemistry,medicine,
sociology,education,and law/history/politics.The corpus thus represents aca-
demic prose as a general register.
The LGSWE presents many frequency findings that describe the patterns of
use for particular grammatical features.Table 1.1 lists the major‘academic fea-
tures’identified in the LGSWE:features that occur much more frequently in
academic prose than in the other registers.For example,corpus analysis in the
LGSWE shows that there are about 300,000 nouns per million words in academic
prose,compared to only around 150,000 per million words in conversation(p.
235).Similarly,there are about 80,000 adjectives per million words in academic
prose versus only 20,000 per million words in conversation(and around 60,000
per million words in fiction and newspapers;p.506).Thus,these grammatical
features can be considered characteristic of academic prose by virtue of their dis-
tribution:although they can be used in all registers,they turn out to be much more
common in academic prose.
Taken together,the findings reported in the LGSWE show that‘academic’
grammatical features come from most structural categories.Three word classes are
especially prevalent:nouns,adjectives,and prepositions.Overall,these grammat-
ical classes are more frequent in academic prose than in other registers,and there
are many related specific features that are especially characteristic of academic
prose(e.g.,nominalizations,noun phrases with multiple modifiers,stance noun
+of-phrase).In contrast,verbs overall are much less common in academic prose
than in other registers,although there are specific verb categories that are typi-
cal of academic prose(e.g.,copula be,existence verbs,derived verbs,and passive
voice verbs).Similarly,there are specific categories of adverbs and adverbials(e.g.,
linking adverbials)that are especially common in academic prose,even though
adverbs overall are more common in spoken registers.
It is difficult to make any global generalizations for dependent clauses.Rather,
specific clause types have their own distributions.For example,finite relative
clauses are much more common in writing than in conversation,but they are most
common in newspaper writing and fiction rather than academic prose.However,
relative clauses with the relative pronoun which are most frequent in academic
prose.Non-finite relative clauses(-ing clauses and-ed clauses)are also especially
common in academic prose(and newspaper writing).Many complement clause
types are actually much more common in conversation than academic prose,es-
pecially that-clauses controlled by verbs(e.g.,I know that he did it).Only oneChapter 1.Introduction
Table 1.1 Grammatical features that are especially common in academic prose(based on a
survey of the Longman Grammar of Spoken and Written English)
Feature Pattern of use
Nouns and noun phrases:
Nouns:overall
p.65
Approximately 60%of all content words in academic prose are
nouns
Nouns vs.pronouns
pp.235–236
Nouns are much more common than pronouns in academic
prose,especially in object positions
Absence of pronouns
pp.235–236
Pronouns are generally rare in academic prose
Specific pronouns:this and
generic one
pp.349–350,354–355
Much more common in academic prose;this is used for
immediate textual reference;one is used for generic rather than
specific reference
Plural nouns
pp.291–292
Much more common in writing than in conversation;most
common in academic prose
Nominalizations
pp.322–323
Much more common in academic prose,especially nouns formed
with-tion and-ity(e.g.distribution,similarity)
Anaphoric expressions
pp.237–238
Anaphoric reference is usually expressed with a determiner+
noun(rather than a pronoun)
Definite article the
pp.267–269
Much more common in writing than in conversation;most
common in academic prose
Demonstrative determiners
pp.270,274–275
Most common in academic prose;especially this and these
Noun phrases with
modifiers
p.578
60%of all noun phrases in academic prose have a modifier
Noun phrases with
pre-modifiers
pp.589,597
Very common in academic prose(and newspapers)
Nouns as pre-modifiers
pp.589–596
Very common in academic prose(and newspapers)(e.g.,
government agencies)
Noun phrases with
post-modifiers
pp.606–608
Very common in academic prose(and newspapers)
Noun phrases with
multiple post-modifiers
pp.640–644
Most common in academic prose(e.g.,the utilization of such
devices for social purposes)
Noun and/or noun
binomial phrases
pp.1033–1034
Most common in academic prose(e.g.,size and shape)
Adjectives and adjective
phrases:
Adjectives:overall
pp.65,506
Adjectives are much more common in academic prose than in
conversation or fiction
Attributive adjectives
pp.506,589
Much more common in academic prose(e.g.,the basic logical
content)??University Language
Table 1.1(continued)
Feature Pattern of use
Specific predicative
adjectives
p.440
Several predicative adjectives are notably more common in
academic prose than in other registers:different,important,
difficult,possible,necessary,available,useful
Derived adjectives
pp.531–533
Much more common in academic prose,especially adjectives
formed with-al(e.g.functional,regional)
Verbs and verb phrases:
Copula be and copular verb
become
pp.359–360,437–439
Most common in academic prose
“Existence”verbs
pp.366,369,419
Much more common in writing than in conversation;Most
common in academic prose(e.g.,include,involve,indicate)
Specific lexical verbs
pp.367–372
Several verbs are notably more common in academic prose than
in other registers:
Activity verbs:use,produce,provide,apply,form,obtain,reduce
Communication verbs:describe,suggest
Mental verbs:consider,assume,determine
Causative/Occurrence/Existence verbs:follow,allow,require,
include,involve,contain,exist,indicate,represent
Specific prepositional verbs
pp.416–418
Several prepositional verbs are notably more common in
academic prose than in other registers:
Activity verbs:deal with,BE applied to,BE used in,BE derived from
Communication verbs:refer to
Mental verbs:BE known as
Causative/Occurrence/Existence verbs:lead to,result in,occur
in,depend on,consist of,BE based on,BE associated with,BE
related to
Verbs with inanimate
subjects
pp.378–380
Common only in academic prose(e.g.,such comparisons
suggest...)
Derived verbs
pp.400–403
Most common in academic prose,especially verbs formed with
re-and-ize(e.g.redefine,computerize)
Tense and aspect
pp.456–462
Academic prose relies primarily on simple aspect,present tense
verb phrases
Passive voice
pp.476–480,937–940
Much more common in academic prose,especially the‘short’
passive(with no by-phrase)
Specific passive verbs
pp.478–480
Several verbs are especially common with passive voice in
academic prose;for example:BE+made,given,taken,used,
found,seen,considered,shown
Adverbs and adverbials:
Specific adverbs
pp.560–563
Several adverbs are notably more common in academic prose
than in conversation:often,usually,significantly,more,relatively,
especially,particularly,generally,indeed
Specific amplifiers
pp.560–563
A few amplifiers are notably more common in academic prose
than in conversation:extremely,highlyChapter 1.Introduction
Table 1.1(continued)
Feature Pattern of use
Specific degree adverbs
pp.566–569
A few degree adverbs are notably more common in academic
prose than in conversation:relatively,fairly,slightly
Linking adverbials
pp.766,880–882
Most common in academic prose;especially
however,thus,therefore,for example(e.g.)
Purpose and concessive
adverbials
pp.784,786,820–826
Most common in academic prose(e.g.,in order to,although)
Dependent clause features:
Relative clauses with the
relative pronoun which
pp.609–612
Most common in academic prose
Participle clauses as
post-modifiers in noun
phrases
pp.606,630–632
Very common in academic prose(and newspapers)(e.g.,the
assumptions given above)
Noun complement clauses
as post-modifiers in noun
phrases
pp.645–655
Very common in academic prose(and newspapers)(e.g.,the fact
that...;the attempt to...)
Abstract noun+of+
ing-clause
pp.653–655
Most common in academic prose,especially with the head nouns
way,cost,means,method,possibility,effect,problem,process,risk
(e.g.,methods of assessing error)
Extraposed that-clauses
pp.672–675
Most common in academic prose,especially controlled by the
adjectives clear,(un)likely,and(im)possible
Extraposed to-clauses
pp.672–675
Most common in academic prose,especially controlled by
adjectives(e.g.,(im)possible,difficult,hard,important,necessary)
Subject predicative
to-clause
pp.714–715,723
Common only in academic prose(and newspapers)(e.g.,The first
step is to evaluate the expression)
ing-clauses controlled by
adjective predicates
p.749
Most common in academic prose;(e.g.,capable of,important
for/in,useful for/in:formalist strategies are useful for analyzing
drama)
Concessive adverbial
clauses
pp.820–825
Most common in academic prose(and newspapers)(though,
although)
Other features
Prepositions
p.92
Most common in academic prose
Of-phrases
pp.301–302
Much more common in writing than in conversation;
Most common in academic prose
Prepositional phrases as
post-modifiers in noun
phrases
p.606–608,634–638
Very common in academic prose(and newspapers)(e.g.,the effect
on the final state)??University Language
Table 1.1(continued)
Feature Pattern of use
Stance noun+of-phrase
pp.984–986
Most common in academic prose,especially possibility of,value of,
importance of,problem of,understanding of
that/those+of-phrase
pp.307–308
Common only in academic prose
Preposition+which in
relative clauses with
adverbial gaps
pp.624–626
Common only in academic prose,especially in which and to which
Selected coordination tags:
and so on,etc.
pp.116–117
Common only in academic prose
Quantifier each Most common in academic prose
Semi-determiners same,
other,certain,and such
pp.282–283
Much more common in academic prose
Dual gender reference:he
or she,his or her,he/she
pp.316–317
Common only in academic prose
Lexical bundles with noun
phrases and/or
prepositional phrases
pp.997,1015–1019
Very common in academic prose;e.g.,the end of the,the nature of
the,one of the most,the way in which,the extent to which,the fact
that the,as a result of,at the time of,in the
case/absence/form/presence of on the basis of,on the other hand
major type of complement clause is especially characteristic of academic prose:
extraposed clauses controlled by adjectives(e.g.,It is possible that...,It is impor-
tant to...).(Subject predicative to-clauses are also found primarily in academic
prose,although they are not especially frequent overall.)Finally,adverbial clauses
are most common overall in fiction.Even conversation uses adverbial clauses to a
slightly greater extent than academic prose.Only one sub-type of adverbial clause
is especially frequent in academic prose:concessive clauses.
In summary,there are few general linguistic features that are uniquely char-
acteristic of academic prose.The most distinctive features of academic prose are
specific grammatical features associated with a particular set of words,such as ex-
traposed complement clauses controlled by stance adjectives.However,a much
larger set of features–such as nouns and prepositional phrases–occur to some
extent in every register;these features can be considered‘academic’because they
are especially common in academic prose.Chapter 1.Introduction
?.?Strengths and weaknesses of previous research
As the survey in the preceding sections show,there have been numerous studies
on academic language over the past two decades.Most of these studies focus on
the use of a particular word or linguistic feature in a particular academic register,
such as the hedge kind of in geology lectures.Taken together,these studies provide
a useful foundation for the study of linguistic similarities and differences among
university registers.
From a structural perspective,previous studies have described features at
many different linguistic levels:words(e.g.,thing),extended collocations and
formulaic language(e.g.,come into play),part-of-speech classes(e.g.,modals),
grammatical/functional classes(e.g.,hedges),syntactic constructions(e.g.,noun
phrase modifiers,conditional adverbial clauses),and overall discourse organiza-
tion.In most studies,linguistic features are described relative to their discourse
functions,such as:conveying informational content;expressing hedging,evalu-
ation,and stance;and signaling the discourse organization(topics,background
context,informational focus or packaging,etc.).In addition,most studies focus on
a particular register and academic discipline,such as biology textbooks or medical
research articles.
Despite the large number of previous studies,there are still some obvious gaps
in what we currently know about university language.First,we lack comprehensive
linguistic descriptions of most university registers.While written academic prose
has been thoroughly investigated,we do not have comprehensive linguistic de-
scriptions for other academic/university registers.Textbooks and academic speech
are probably the best studied of these other registers,with several studies describ-
ing the functions of particular linguistic features.Taken together,these studies
identify some of the salient features of those registers,but they do not amount
to a comprehensive linguistic description of either textbooks or academic speech.
And we know even less about registers like office hours,study groups,and institu-
tional writing(e.g.,university catalogs),because there have been few studies of any
kind based on texts from those registers.In addition,apart from previous Multi-
Dimensional studies,we know little about the overall patterns of variation among
university registers.The T2K-SWAL Project,introduced in the following section,
was designed to help fill these gaps.
?.?Background of the present book:Introduction to the T2K-SWAL Project
The present book grew out of the TOEFL 2000 Spoken and Written Academic
Language(T2K-SWAL)Project,sponsored by the Educational Testing Service(see
Biber et al.2004).The project included three major components:??University Language
–the construction of a large corpus of spoken and written university registers
(the T2K-SWAL Corpus);
–the description of language use in the university based on extensive linguistic
analysis of that corpus;
–the development of analytical tools to describe the linguistic characteristics of
exam prompts relative to the corpus.
The initial motivation for the project grew out of the need for an external standard
to evaluate the representativeness of ESL/EFL materials.That is,given the lack of
a comprehensive description of university language,it has been nearly impossible
to evaluate the extent to which textual materials for ESL/EFL instruction and as-
sessment actually represent the language of the target registers.Specifically in the
context of the TOEFL 2000 effort undertaken by ETS(see Jamieson et al.2000),
we lacked the tools to determine whether the texts used in listening and reading
portions of the TOEFL exam accurately represent the linguistic characteristics of
spoken and written academic registers(see Enright et al.2000;Bejar et al.2000).
One reason for this gap is that there have been no readily available text corpora
of university language that could be used for research studies.The T2K-SWAL
Corpus was compiled to meet this need.Then,based on analysis of that corpus,
the T2K-SWAL project focused on two primary goals:
–to identify the salient patterns of language use found in each university register
(and across disciplines,levels,etc.);
–to develop diagnostic tools and procedures for assuring that the language used
in TOEFL Listening Comprehension and Reading Comprehension tasks were
representative of actual language use in the university.
Biber et al.(2004)is a technical report describing the T2K-SWAL Project and the
design,construction,and linguistic analysis of the T2K-SWAL Corpus.
The T2K-SWAL Project was a large-scale study,involving the collaborative ef-
forts of multiple researchers.Although the project was coordinated at Northern
Arizona University,and most linguistic analyses were carried out there,the col-
lection of texts for the corpus was supervised by four co-principal investigators
at different universities:Susan Conrad at Iowa State University;Randi Reppen
at NAU;Pat Byrd at Georgia State University;and Marie Helt at California State
University,Sacramento.In addition,there were numerous research assistants and
student workers from all four universities who helped collect,transcribe,scan,tag,
tag-edit,and analyze the corpus.
Work on the T2K-SWAL Project proceeded in three major stages:(1)design,
construction,and grammatical“tagging”of the T2K-SWAL Corpus;(2)linguistic
analysis of the patterns of register variation in the corpus;and(3)development of